Last active
September 8, 2015 15:28
-
-
Save biomadeira/d2c9d27407c8bf4e90d8 to your computer and use it in GitHub Desktop.
Modified perl script for updating BIOLIP
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #!/usr/bin/perl -w | |
| use strict; | |
| # BIOLIP from Zhang Lab | |
| # http://zhanglab.ccmb.med.umich.edu/BioLiP/ | |
| # Folder Structure | |
| # ================ | |
| # NOBACK/ | |
| # DB/ | |
| # biolip/ | |
| # database/ | |
| # biolip.txt | |
| # Get old datasets: | |
| # wget http://zhanglab.ccmb.med.umich.edu/BioLiP/download/BioLiP.tar.bz2 | |
| # tar -zxvf BioLiP.tar.bz2 | |
| # mv BioLiP_2013-03-6.txt biolip_all_data.txt | |
| # then run (modified) download_all_sets.pl to update and then | |
| # cat database/biolip_all_data.txt | cut -f 1,2,5,6 > biolip.txt | |
| # `mkdir -p BioLiP_updated_set`; | |
| #`rm -fr BioLiP_updated_set/*`; | |
| chdir "database"; | |
| `rm weekly.html`; | |
| my $head= "http://zhanglab.ccmb.med.umich.edu/BioLiP/weekly"; | |
| my $address="http://zhanglab.ccmb.med.umich.edu/BioLiP/weekly.html"; | |
| system("wget -o log -c $address") == 0 or die "System call failed: $!"; | |
| my @rst=`cat weekly.html`; | |
| my @all=(); | |
| foreach my $r(@rst) | |
| { | |
| if($r =~ /\<tr\>\<td\>(\S+)\<\/td\>/) | |
| { | |
| # print "$1\n"; | |
| push(@all, $1); | |
| } | |
| } | |
| my $tot=@all; | |
| print "\n====================================================\n"; | |
| print "In total, there are $tot weeks to update.\n\n"; | |
| my $annotation="biolip_all_data.txt"; | |
| # open(OUT, ">$annotation"); | |
| # close(OUT); | |
| # my $annotation1="BioLiP_UP_nr.txt"; | |
| # open(OUT, ">$annotation1"); | |
| # close(OUT); | |
| foreach my $r(@all) | |
| { | |
| my $rec="receptor_$r.tar.bz2"; | |
| my $rec1="receptor1_$r.tar.bz2"; | |
| my $lig="ligand_$r.tar.bz2"; | |
| my $ano="BioLiP_$r.txt"; | |
| my $ano1="BioLiP_$r\_nr.txt"; | |
| if(-s $ano) | |
| { | |
| print "The week $r was updated before, skip this one\n"; | |
| } | |
| else | |
| { | |
| print "Dowload redundant set for the week $r...\n"; | |
| # system("wget -o log -c $head/$rec") == 0 or die "System call failed: $!"; | |
| # system("tar -xvf $rec >log")== 0 or die "System call failed: $!"; | |
| # system("wget -o log -c $head/$rec1") == 0 or die "System call failed: $!"; | |
| # system("tar -xvf $rec1>log")== 0 or die "System call failed: $!"; | |
| # system("wget -o log -c $head/$lig") == 0 or die "System call failed: $!"; | |
| # system("tar -xvf $lig >log")== 0 or die "System call failed: $!"; | |
| system("wget -o log -c $head/$ano") == 0 or die "System call failed: $!"; | |
| # print "Dowload non-redundant set for the week $r...\n"; | |
| # $rec="receptor_$r\_nr.tar.bz2"; | |
| # $rec1="receptor1_$r\_nr.tar.bz2"; | |
| # $lig="ligand_$r\_nr.tar.bz2"; | |
| # system("wget -o log -c $head/$rec") == 0 or die "System call failed: $!"; | |
| # system("tar -xvf $rec >log")== 0 or die "System call failed: $!"; | |
| # system("wget -o log -c $head/$rec1") == 0 or die "System call failed: $!"; | |
| # system("tar -xvf $rec1>log")== 0 or die "System call failed: $!"; | |
| # system("wget -o log -c $head/$lig") == 0 or die "System call failed: $!"; | |
| # system("tar -xvf $lig >log")== 0 or die "System call failed: $!"; | |
| # system("wget -o log -c $head/$ano1") == 0 or die "System call failed: $!"; | |
| } | |
| `cat $ano >> $annotation`; | |
| # `cat $ano1 >> $annotation1`; | |
| #last; | |
| } | |
| print "Cheers! All updates are done.\n"; | |
| print "====================================================\n\n"; | |
| # print "Please download the old sets manually at | |
| # http://zhanglab.ccmb.med.umich.edu/BioLiP/download.html | |
| # Please read http://zhanglab.ccmb.med.umich.edu/BioLiP/download/readme.txt | |
| # about explanation of the annotation file. | |
| # "; | |
| # print "Please feel free to contact me (Jianyi, yangji\@umich.edu) if you have any problems with BioLiP.\n | |
| # Thanks for using the BioLiP database! | |
| # --------------------------------------------- | |
| # Please cite the following paper if you use BioLiP in your projects: | |
| # Jianyi Yang, Ambrish Roy, Yang Zhang, BioLiP: a semi-manually curated database for biologically | |
| # relevant ligand-protein interactions, Nucleic Acids Research, 41:D1096-D1103, 2013. | |
| # "; | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment